A conventional type visible light semiconductor light emitting device employing, for instance, In0.49(Ga1-zAlz)0.51P based compound semiconductor may be arranged as exemplarily illustrated in FIG. 4. That is, FIG. 4 illustrates a light emitting layer forming portion 29 of double hetero structure in which there are deposited, on a semiconductor substrate 21 of n-type GaAs, an n-type clad layer 22 made of a n-type InGaAlP based semiconductor material, an active layer 23 made of a InGaAlP based semiconductor material having a composition with which the band gap energy becomes smaller than that of the clad layers, and a p-type clad layer 24 of a p-type InGaAlP based semiconductor material, respectively, through epitaxial growing. A p-type window layer (current diffusing layer) 25 made of an AlGaAs based compound semiconductor is further deposited on a surface thereof. Moreover, a p-side electrode 27 is formed on a central portion of this surface with a contact layer 26 of GaAs being interposed between them while an n-side electrode 28 is formed on a rear surface of the semiconductor substrate.
The window layer 25 is employed for the purpose of achieving two functions, namely enabling easy extraction of light from side surfaces and enabling easy light emission at the light emitting layer forming portion on the entire surface of the chip by diffusing current for spreading the same over the entire surface of the chip, and it is formed by a semiconductor layer of small light absorption rate and large carrier density.
In such a semiconductor light emitting device, light directed towards the p-side electrode 27 that is provided on the surface side will be shielded by this electrode 27 so that no light can be extracted to the surface side. For eliminating such loss, it is known to provide an arrangement in which a semiconductor layer of different conductivity from the conductivity of the adjoining semiconductor layers or an insulating layer is interposed between any of the semiconductor layers, which is underlying the p-side electrode 27 as indicated by 31 in FIG. 5 so as to restrict current to directly underneath the p-side electrode 27 and thus to avoid emitting light directly underneath thereof.
As already discussed, in a conventional semiconductor light emitting device in which light is extracted from a surface side of laminated semiconductor layers, light emitted in the interior cannot be sufficiently extracted since light is shielded by an electrode that is provided on the surface side. Such an electrode on the surface side requires a bonding area of approximately 80 to 100 μmφ for wire bonding or the like, in a size of the chip area of approximately 200 to 300 μm square, which leads to a drawback in that most of the area is shielded by the electrode so as to degrade the efficiency of extraction of light to the exterior.
Further, even though current is prevent from flowing by providing a current blocking layer downward of the upper electrode (p-side electrode) as illustrated in the above-mentioned FIG. 5, current will still flow to the active layer portion further downward thereof to cause emission of light, and even if emission of light has been successively been prevented, light emitted in the periphery thereof will be absorbed by the active layer, which is a portion that does not emit light, so that portions that emit light downward of the upper electrode cannot be effectively used.
Moreover, when trying to extract light from sidewall of the chip, the side wall will only be the periphery of the rectangular chip while the light emitting region is, in the presence of a region in which current injection is blocked, substantially the entire surface of the chip area of the active layer that is formed to be vertical with respect to the side wall excluding the above blocking region, and light emitted in the interior such as the central portion of the chip cannot be effectively extracted to the exterior owing to factors such as absorption by the semiconductor layers.